Adjunctive Corticosteroid Use and Clinical Outcomes in Non-HIV Pneumocystis jirovecii Pneumonia.

Abstract

Adjunctive corticosteroids improve outcomes in HIV-associated Pneumocystis jirovecii pneumonia (PCP), but their role in non-HIV-associated PCP is uncertain. Prior evidence largely has been limited to binary treatment groups and rarely has accounted for daily or cumulative dose effects. What is the dose-response relationship between adjunctive corticosteroids and outcomes in non-HIV immunocompromised adults with PCP requiring supplemental oxygen? We conducted a multicenter retrospective cohort analysis of 375 non-HIV immunocompromised adults with proven or probable PCP hospitalized between 2019 and 2025. All patients were hypoxemic at treatment initiation. Corticosteroid exposure was modeled as a continuous, cumulative time-varying dose over 21 days using marginal structural models with inverse probability of treatment weighting to adjust for baseline covariates and time-varying illness severity. Of 375 patients, 351 patients (93.6%) received corticosteroids. The most common causes of immunosuppression were hematologic malignancy (30%), solid tumors on chemotherapy (30%), autoimmune disease (17%), and solid organ transplantation (14%); 56% of patients required ICU admission and 44% of patients died within 90 days. Greater cumulative steroid dose was associated with increased risk of 90-day mortality (weighted hazard ratio [HR], 1.01 per 100 mg prednisone-equivalent; 95% CI, 1.00-1.02; P = .006). Steroid exposure was not associated with risk of intubation (HR, 0.99; 95% CI, 0.97-1.02) or faster liberation from advanced respiratory support (HR, 1.00; 95% CI, 0.98-1.02). In patients with non-HIV PCP with hypoxemia, higher cumulative corticosteroid exposure was not associated with improved respiratory outcomes and was linked to increased mortality. Use of doses exceeding trial-tested regimens should be approached with caution.